The Group Engagement Model: Procedural Justice, Social Identity, andCooperative Behavior

Tom R. TylerDepartment of Psychology

New York University

Steven L. BladerStern School of Business

New York University

The group engagement model expands the insights of the group-value model of pro-cedural justice and the relational model of authority into an explanation for whyprocedural justice shapes cooperation in groups, organizations, and societies. It hy-pothesizes that procedures are important because they shape people’s social iden-tity within groups, and social identity in turn influences attitudes, values, and be-haviors. The model further hypothesizes that resource judgments exercise theirinfluence indirectly by shaping social identity. This social identity mediation hy-pothesis explains why people focus on procedural justice, and in particular on pro-cedural elements related to the quality of their interpersonal treatment, becausethose elements carry the most social identity-relevant information. In this article,we review several key insights of the group engagement model, relate these insightsto important trends in psychological research on justice, and discuss implicationsof the model for the future of procedural justice research.

The original goal of social justice research wasto demonstrate the power of justice judgments toshape people’s thoughts, feelings, and actions (Ty-ler, 2000; Tyler, Boeckmann, Smith, & Huo, 1997;Tyler & Smith, 1997). Justice studies have, in fact,continually provided strong and consistent demon-strations of support for this basic justice hypothe-sis. Justice has an impact; it is substantial in magni-tude; it is consistently found across a wide varietyof group and organizational contexts; and it is dis-tinct from judgments of self-interest or per-sonal/group gain. This conclusion suggests that in-formation about justice is central to people’sevaluations of social situations (Tyler et al., 1997;van den Bos & Lind, 2002).

Justice research has evolved a great deal in theprocess of developing these insights about the roleof justice in social contexts. Numerous models re-lated to the justice phenomena have been proposed.Some of these continue to shape the face of justiceresearch today, whereas others have fallen into rela-tive obscurity by their inability to withstand empiri-cal scrutiny. All have contributed to the history ofjustice research.

In this article, we put forth a theoretical modelthat develops from the findings of earlier models andshifts in the focus of justice research. Specifically,we present theory and research on our group engage-ment model, which draws together the insights of thegroup-value model of procedural justice (Lind & Ty-ler, 1988) and the relational model of authority (Ty-ler & Lind, 1992) and extends them to understandthe antecedents of cooperation in groups (Tyler &Blader, 2000). The argument underlying the groupengagement model is that people’s focus on proce-dural justice sheds light on their motivations for en-gaging in groups, and thus the model explicitly pos-its what those motivations are. In so doing, itcontributes to our understanding of what people areseeking when they involve themselves in groups andthe importance of justice in social settings. Themodel also suggests some innovative directions forfuture research.

Because it is important to understand the past toevaluate new theories and models, we preview ourpresentation of the model by a discussion of majorshifts in justice research and how they are addressedby the group engagement model. We also explicitlydistinguish the group engagement model from earliermodels that contributed to its development, and thenprovide a more in-depth treatment of the model andthe propositions it raises for future research.

349

Personality and Social Psychology Review2003, Vol. 7, No. 4, 349–361

Copyright © 2003 byLawrence Erlbaum Associates, Inc.

Requests for reprints should be sent to Tom R. Tyler, Depart-ment of Psychology, New York University, 6 Washington Place,Room 550, New York, NY 10003. E-mail: tom.tyler@nyu.edu

The History of Social Justice Research

The Shift From Distributive toProcedural Justice

Early research on justice focused on the argumentthat people’s feelings and behaviors in social interac-tions flow from their assessments of the fairness oftheir outcomes when dealing with others (distributivefairness). This hypothesis was widely supported. Inparticular, experimental studies showed that peoplewere most satisfied when outcomes were distributedfairly (Walster, Walster, & Berscheid, 1978). What wasmost striking and provocative about these results werethe adverse reactions by those who received more thanthey felt they deserved; people did not react well to be-ing ‘over-benefited.’This finding suggested that peoplewill give up resources and accept less when they be-lieve doing so is fair.

Despite the impressive findings of early studies ofdistributive justice, the focus of attention among justiceresearchershas increasinglyshiftedawayfromstudyingonly distributive justice to a focus on people’s distribu-tive and procedural justice concerns. A number of fac-tors have driven this shift. First, research shows that dis-tributive justice judgments are often biased (e.g.,Messick & Sentis, 1985; Ross & Sicoly, 1979; Thomp-son & Loewenstein, 1992). This limits the utility of dis-tributive justice as a construct, because people will oftensee themselves as deserving more favorable outcomesthan others see them as deserving. As a result, peoplefrequently cannot be given what they feel they deserve,and distributive justice has not proven as useful in re-solving group conflicts as was initially hoped.

A greater focus on procedural justice issues wasalso driven by later studies that looked simultaneouslyat the impact of distributive and procedural justicejudgments and found a predominant influence of pro-cedural justice on people’s reactions in groups (Alex-ander & Ruderman, 1987; Tyler & Caine, 1981). Thesestudies, conducted in settings in which people had in-formation about both distributive and procedural jus-tice, found that procedural justice judgments play themajor role in shaping people’s reactions to their per-sonal experiences. More recent research echoes thesefindings about the relative impact of procedural anddistributive justice concerns (Tyler & Blader, 2000). Inaddition, people who were asked to talk about personalexperiences of injustice were found to talk primarilyabout procedural issues, in particular about beingtreated with a lack of respect when dealing with others(Messick, Bloom, Boldizar, & Samuelson, 1985;Mikula, Petri, & Tanzer, 1990).

Justice research has followed the path outlined bythis evidence because it finds that the primary impacton people comes from their judgments about the fair-ness of procedures (see Tyler, Boeckmann, Smith, &

Huo, 1997; Tyler & Smith, 1997). This does not mean,of course, that people no longer study distributive jus-tice, but that there is a particularly strong focus in cur-rent research on issues of procedural justice. This focusis embodied in the group engagement model by the keyrole it accords to procedural justice.

The Focus on Treatment Issues inDefinitions of Procedural Justice

A second important shift in justice research hasbeen a change in how procedural justice is defined.Early work on procedural justice was guided by the in-fluential research program of Thibaut and Walker(1975). Thibaut and Walker centered their proceduraljustice studies on procedures as mechanisms for mak-ing decisions about the allocation of outcomes. In par-ticular, they focused on formal procedures that relatedto decision-making processes in legal settings. SoThibaut and Walker linked their discussions of proce-dures primarily to issues of decision making, and inparticular to issues of decision making about allocationdecisions. Because their procedural models wererooted in an era in which distributive justice domi-nated, their focus was natural. This context also influ-enced their theory development, because they linkedpeople’s desire for fair procedures to their desire toachieve equitable outcomes. They proposed that peo-ple value procedural justice (operationalized in theirresearch as voice or process control) because it facili-tates decision makers’ ability to make equitable judg-ments. In other words, procedures are valued insofar asthey affect the outcomes that are associated with them.

This focus on decision making in allocation con-texts is no longer true of procedural justice research.Researchers have increasingly moved their attentionaway from an exclusive focus on the decision-makingfunction of procedures to include more attention to theinterpersonal aspects of procedures. Those interper-sonal aspects of procedures arise because proceduresare settings within which people are involved in a so-cial interaction with one another. This is true regardlessof whether the procedure involves bargaining, a marketexchange, team interaction among equals, or a thirdparty procedure with a decision maker, such as media-tion or a trial.

In social interactions there is considerable variationin the manner in which people treat one another. Theycan act politely, rudely, respectfully, with hostility, andso on. These aspects of the interpersonal experience ofa procedure—which occur in the context of an interac-tion whose overt purpose is to make a decision to allo-cate resources or resolve a conflict—may also influ-ence those who are involved.

An example of this shift from an exclusive focus ondecision making to a focus that includes attention to theinterpersonal quality of the interaction can be found in

350

TYLER AND BLADER

the literature on voice or process control. In the earlywork of Thibaut and Walker (1975), the opportunity topresentevidencewas linked to thedesire to influence thedecisions made by third party decision makers. Thevalue of the opportunity to speak was directly related totheir estimate of how much influence they had over thedecision maker. Consequently, in this research peoplewere not asked about whether they were treated politelyand with dignity by the decision maker.

However, later studies of voice suggested that havingthe opportunity for “voice” had interpersonal or“value-expressive” worth that was not linked to any in-fluence over the decisions made (Tyler, 1987). Thesestudies showed that people still rated a procedure to bemore fair if they had voice, even if they knew that whatthey said had little or no influence on the decisions made(Tyler, Rasinski, & Spodick, 1985). This was true evenwhen the opportunity for voice came after the decisionwasalreadymade(Lind,Kanfer,&Earley,1990).Thesefindings suggest that voice has value beyond its abilityto shape decision-making processes and outcomes.

What factors are driving the influence of voice, evenwhen it clearlycannotaffect theeventualoutcomeorde-cision? If an authority listens to people’s arguments, wemight hypothesize that people think that the authority isconferring interpersonal respect on that person. This ar-gument was supported by the finding that people onlyvalue such voice opportunities if they feel that the au-thority is “considering” their arguments (Tyler, 1987).This suggests that people were focused on whether ornot they had their concerns and needs in the situationtreated respectfully by the decision maker, independ-ently of whether or not the course of action they recom-mend to resolve those concerns was adopted.

Other research on people’s procedural justice con-cerns directly measured people’s focus on the qualityof their interpersonal treatment (“standing” or “statusrecognition”), and found that it had an effect that wasdistinct from their interest in the fairness of deci-sion-making judgments. Drawing on these findings,the relational model of authority (Tyler & Lind, 1992)explicitly included issues of interpersonal treatmentwithin the framework of procedural justice concerns.The relational model, therefore, directly recognizedthe importance of interpersonal treatment. Subsequentstudies confirm that issues of interpersonal treatmentor standing independently shape procedural justicejudgments (Tyler, 1988, 1994; Tyler & Huo, 2002).

These interpersonal aspects of procedures have beenfound by recent studies to be so powerful in their impactthat some researchers have argued that they might po-tentially be treated as a separate type of “interactional”justice (Bies & Moag, 1986; Tyler & Bies, 1990). Re-gardless of whether the quality of the treatment that peo-ple experience via procedures is actually considered adistinct form of justice (see Blader & Tyler, 2003a,2003b), justice researchers have again followed their

findings about what impacts the people they study. Thishas led them to increasingly turn their research towardexploring interpersonal or interactional aspects of pro-cedures—which are reflected in judgments about thequality of one’s treatment by others.

The group engagement model not only incorporatesthis shift in the focus of how justice is defined—by in-corporating quality of treatment issues—but also pro-vides a framework for understanding why this class ofprocedural criteria has the impact that it does.

Moving From Anger and NegativeBehaviors to Positive Attitudes–Valuesand Cooperative Behaviors

Early research on justice was rooted in the literatureon relative deprivation, a literature whose origins lie inefforts to understand and explain riots and rebellion(Crosby, 1976; Gurr, 1970). This focus on negative at-titudes and behaviors continued in later efforts to un-derstand distributive influences on pay dissatisfaction,employee theft, sabotage and turnover, and proceduraleffects on resistace to third-party decisions (Tyler &Smith, 1997). However, recent research on proceduraljustice has increasingly focused on more prosocial out-comes, such as how to build trust, encourage responsi-bility and obligation, generate intrinsic motivation andcreativity, and stimulate voluntary cooperation withothers (Tyler & Blader, 2000). Similarly, there hasbeen increasing attention to exploring when justicemotivations encourage people to provide resources tothe disadvantaged (Montada, 1995). Interestingly, thisshift is consistent with a shift that has been taking placewithin psychological research more generally (Snyder& Lopez, 2002).

This broadening of the focus of justice research isconsistent with the group engagement argument thatjustice theories provide a basis for understanding peo-ple’s general relationship to groups. That includes bothpeople’s negative reactions to injustice and the abilityof justice to motivate engagement and cooperation. So-ciety, after all, does not just want people not to riot ordestroy. It also wants them to be happy, creative, andproductive.

Models of the Psychology of Justice

While continually supporting the basic importanceof people’s justice judgments, these shifts in focus haveresulted in a dramatic change in the character of justiceresearch since the 1960s. In fact, early justice research-ers might have trouble recognizing many recent justicestudiesasbeingabout justice—at least as theyoriginallyunderstood that construct. Instead of viewing justice asresiding in the rules used in the distribution of resourcesin a group, justice is more recently viewed as beingstrongly linked to quality of treatment issues, such as

351

JUSTICE, IDENTITY, AND BEHAVIOR

treating people with politeness and dignity in social in-teractions. It is also focused on stimulating commitmentand cooperation, rather than minimizing anger and de-structive behaviors.

We argue that these changes—which were guidedby the empirical results of justice research—can bestbe understood by considering the psychological dy-namics underlying justice. That is, they can be ex-plained by considering the psychological processesthat lead people to react to issues of justice or injusticewhen they are dealing with others. Much early justiceresearch was focused on showing that justice matters,that is, on demonstrating that people’s thoughts, feel-ings, and behaviors are shaped by their justice judg-ments, suggesting that information about justice is cen-tral to people’s evaluations of social situations (Tyler etal., 1997; van den Bos & Lind, 2002). However, to de-velop a deeper understanding of why these effectsemerge—and why the shifts in research focus we haveoutlined have occurred—we need to pay attention tothe psychology underlying justice.

Several models have been proposed to understand thepsychology underlying procedural justice. We will be fo-cusing here on a set of models that share an emphasis onthe relational implications of justice evaluations. Thesemodels represent a significant systematic research pro-gram designed to understand the psychology of justice. Inparticular, we will be presenting our group engagementmodel, which integrates the insights of the earlier groupvalue (Lind & Tyler, 1988) and relational models (Tyler& Lind, 1992) and extends those insights into an explana-tion for why procedural justice shapes cooperation ingroups, organizations, and societies. The relationshipamong these three models is shown in Table 1.

The models differ first in their focus. Thegroup-value model focuses on the antecedents of judg-ments of procedural justice. The relational model ex-plores the factors shaping reactions to authorities. Themodels also differ in their predictions. The group-valuemodel predicts that noninstrumental factors will influ-ence procedural justice judgments, a prediction con-firmed both by findings of noninstrumental voice ef-fects (Lind, Kanfer, & Earley, 1990; Tyler, 1987), andby demonstrations that people care more about issues

of procedural justice when dealing with members oftheir own groups (Tyler, 1999). The relational modelpredicts that procedural justice will influence reactionsto authorities, as has been subsequently found by stud-ies of legal, political, managerial, familial, and educa-tional authorities (Tyler & Smith, 1997). It further pre-dicts that relational concerns—in particular neutrality,trustworthiness, and status recognition—will influ-ence procedural justice judgments, an argument sup-ported by a number of studies (Tyler, 1989, 1994; Ty-ler, Degoey, & Smith, 1996).

How does the group engagement model differ fromthese earlier, empirically supported models? First, thegroup engagement model is broader in its scope. Theobjective of the model is to identify and examine theantecedents of attitudes, values, and cooperative be-havior in groups. Hence, the group engagement modelbroadens the focus of justice studies and its predeces-sor models of justice by positing a general model of therelationship between people and groups. In trying tounderstand the precursors of people’s engagement intheir groups, it identifies and examines a much broaderset of variables—and dynamics between those vari-ables—than earlier justice models.

Second, several new ideas and hypotheses flowfrom the group engagement model. It predicts thatidentity judgments will be the primary factors shapingattitudes, values, and cooperative behaviors in groups.Second, it predicts that resource judgments will moststrongly influence attitudes, values, and discretionarycooperative behaviors in groups through their indirectinfluence on identity judgments, rather than directly.Third, it predicts that the primary antecedent of iden-tity judgments will be judgments about the proceduraljustice of the group. Fourth, it predicts that status judg-ments about pride and respect will shape identificationwith the group. Each of these novel predictions is elab-orated on in the next section.

The Group Engagement Model

As noted, the key objective of the group engage-ment model is to understand what shapes the relation-

352

TYLER AND BLADER

Table 1. Comparison of Models of Procedural Justice

Model Focus of Concern Value Added by Model

Group-value model Procedural justice judgments Noninstrumental factors influence judgments about procedural justiceRelational model Authority relations, leadership Procedural justice shapes reactions to authorities

Relational concerns (neutrality, trustworthiness, standing and statusrecognition) shape judgments about procedural justice

Group engagement model Attitudes, values, and cooperativebehavior in groups

Identity judgments directly shape attitudes, values, and cooperative behaviorResource judgments influence attitudes values, and discretionary

cooperative behavior primarily through their influence on identityjudgments

Procedural justice shapes identity judgmentsPride and respect influence identification with the group

ship that people form with their groups. People haveconsiderable discretion about the degree to which theyinvest themselves in their groups by working on behalfof the group. To examine this issue, the group engage-ment model distinguishes between two classes of co-operative behavior: mandatory and discretionary.Mandatory cooperation is behavior that is stipulated bythe group, whereas discretionary cooperation origi-nates with the group member. The model argues thateach of these forms of cooperation is differently moti-vated. Of the two, mandatory behaviors are morestrongly affected by incentives and sanctions, becausethey are behaviors required by the group and thus thegroup specifically structures incentives and sanctionsto encourage these behaviors. Discretionary behaviorsare more strongly under the influence of people’s inter-nal motivations (their attitudes and values), becausethey are behaviors that originate with the individual.Because discretionary cooperative behaviors are espe-cially valuable to groups (see Tyler & Blader, 2000),the precursors of such behavior are especially centralto discussions of the motivation of group members.

Further, people do not only have leeway with howthey act on behalf of their groups. They also have dis-cretion in the degree to which they hold positive atti-tudes and values toward the group. Attitudes can serveas internal motivations that encourage people to en-gage in behaviors that benefit the group and that theyfind personally rewarding (i.e., behaviors they “want”to do). Values are feelings of responsibility that shapepeople’s sense of behaviors that they should do (i.e.,behaviors they “ought” to do), and can thus also serveas internal motivations. For instance, they may reflectfeelings of responsibility and obligation to followgroup rules and the orders of group leaders. They canbe rooted in either their moral values or their viewsabout the legitimacy of group rules and authorities,both of which are at the discretion and control of the in-dividual group member.

Both attitudes and values are important becausethey lead people to be internally motivated to engage inand cooperate with the group. To the degree that peopleare internally motivated, they engage in cooperativebehaviors for personal reasons, and they do not need toreceive incentives (rewards) or to face the risk of sanc-tions (punishments) to encourage their group-relatedbehaviors. This benefits groups, which are then free todeploy their assets in other ways that benefit the group.

Understanding how people negotiate this latitude inhow they think, feel, and act toward the group is impor-tant for understanding the psychology that drives peo-ple in group settings. To address this issue, the groupengagement model contrasts two social psychologi-cal–motivational models concerning the reasons thatpeople have for engaging in groups. These models arethe resource-based social exchange model (Kelley &Thibaut, 1978; Rusbult & Van Lange, 1996; Thibaut &

Kelley, 1959) and the identity-based social identitymodel (Hogg & Abrams, 1988).

The group engagement model argues that groupsbenefit when the people within them engage them-selves in the group, and groups are particularly benefit-ted when that engagement is based on internal motiva-tions because cooperation does not then depend on theability of the group to utilize incentives or sanctions.This leaves open the question of how to best encouragesuch internal motivation. We address that issue in thenext section, in which we compare the two social psy-chological–motivational models outlined previouslyand their linkages to cooperation.

The Influence of Identity and ResourceMotivations on Engagement in Groups

The group engagement model contrasts two poten-tially important aspects of groups, either or both ofwhichmight shapegroupmember’scooperationanden-gagement: the group’s identity implications for the per-sonwithin thegroup (Hogg&Abrams,1988)and the re-sources that the person gains and loses from groupmembership, in either absolute terms or relative to whatis available in other groups (Thibaut & Kelley, 1959).

The group engagement model selects among thesetwo aspects of groups and argues that the central reasonthat people engage themselves in groups is because theyuse the feedback they receive from those groups to cre-ate and maintain their identities. In other words, thegroup engagement model hypothesizes that, of the twotypes of motivations, it is the development and mainte-nance of a favorable identity that most strongly influ-ences cooperation. The model predicts that people’swillingness to cooperate with their group—especiallycooperation that is discretionary in nature—flows fromthe identity information they receive from the group.That identity information, in turn, is hypothesized toemanate from evaluations of the procedural fairness ex-perienced in thegroup.This suggests that identityevalu-ations and concerns mediate the relationship betweenjustice judgments and group engagement. We will referto this as the social identity mediation hypothesis.

Why might this be so? Using social identity theoryas our framework, we argue that an important functionof groups is to provide people with a way of construct-ing a social identity. It is widely recognized that groupsshape people’s definitions of themselves and their feel-ings of well-being and self-worth (Hogg & Abrams,1988; Sedikides & Brewer, 2001). In particular, groupmemberships shape people’s conceptions of their so-cial selves—the aspect of the self that is formedthrough identification with groups. Groups help to de-fine who people are and help them to evaluate their sta-tus. The first part of this process involves social catego-rization, the taking on of the categories that defineone’s group and using them to construct one’s self-im-

353

JUSTICE, IDENTITY, AND BEHAVIOR

age. The second part of the process involves linkingviews about self-worth and self-esteem to group mem-berships. Thus, to some degree people’s sense of theirown worth is linked to the groups to which they belong.This aspect of self, as opposed to the personal self(unique individual traits), or the relational self (the selfdefined by dyadic relationships), will be the focus ofthis discussion. Sedikides and Brewer (2001) referredto the aspect of the self we consider as the collectiveself—the self linked to group memberships.

The group engagement model distinguishes amongthree aspects of group-linked or social identity: identi-fication, pride, and respect. Identification reflects thedegree to which people cognitively merge their senseof self and their evaluations of self-worth with theirjudgments of the characteristics and status of theirgroups. Pride reflects the person’s evaluation of the sta-tus of their group. Respect reflects their evaluation oftheir status within the group. The group engagementmodel argues that each of these aspects of identityplays an important role in people’s relationship to theirgroup. This argument is shown in Figure 1. It focuseson what people get from groups in the form of ac-knowledgment and recognition of their identities.

As we have noted, the identity-based model of co-operation can be contrasted to a resource-based modelof cooperation. Social psychologists have long recog-nized that people interact with others to exchange ma-terial resources. These material resources can varywidely—from things such as food to money—but re-gardless they share the characteristic of being materialresources that people obtain through their cooperationwith others (Thibaut & Kelley, 1959). Guided by social

exchange models, many social psychological discus-sions of people’s relationships with groups haveargued that this exchange of material resources is thefundamental reason that people engage in groups.

The social exchange perspective is the basis of sev-eral more recent resource-based models, including theinvestment model, which focuses on exit and loyalty togroups (Rusbult & Van Lange, 1996); realistic groupconflict theory (Levine & Campbell, 1972); models ofleader–member exchange (Dansereau, Graen, & Haga,1975); models of in-role behavior (Eisenberger, Hun-tington, Hutchinson, & Sowa, 1986); goal theories ofmotivation (Locke & Latham, 1990); and sanc-tion-based models of regulation (Nagin, 1998).

These resource-based perspectives predict that peo-ple’s level of cooperation with a group will be shaped bythe level of the material resources that they receive fromthat group and the sanctioning risks they face within thegroup. Thus, the willingness to voluntarily cooperatewith the group by doing things that help the group flowsfromassessmentsof thedesirabilityof the resources thatare gained or lost by association with the group. In addi-tion, loyalty to the group will also be shaped by the levelof resources people are obtaining, relative to what theymight obtain in another group (Rusbult & Van Lange,1996). Though social exchange perspectives by andlarge emphasize the importance of material resources,such as food, money, security, and so on, it is importantto note that some models acknowledge that individualsmay develop a long-term perspective on resource ex-change (Foa & Foa, 1974; Kelley & Thibaut, 1978).

The group engagement model proposes that theidentity model prevails over the resource model in pre-

354

TYLER AND BLADER

Figure 1. The group engagement model.

dicting engagement and cooperation. It argues that re-source judgments do not directly shape engagement.This is not to say that the group engagement model ar-gues that resource judgments have no influence on en-gagement. Instead, the model hypothesizes that re-source judgments indirectly influence most forms ofengagement by shaping identity. That is, to some de-gree, people evaluate their identity and status in a par-ticular group by the level of the resources that they arereceiving from that group. To the extent that havingmore resources in a group leads people to feel betterabout their identity with the group, they will engagethemselves more in that group.

It is not obvious that people’s engagement in groupswould be the result of identity judgments. People couldpotentially consider a wide variety of aspects of theirrelationship to their group when they are evaluating thedegree to which they want to engage themselves in agroup. One thing that we might expect people to con-sider is reward level—that is, people might considertheir salaries; the number of resources they are given tomanage; and the size of their office, their car, or theirhome as key inputs into their judgments about howmuch to engage themselves in a group. The group en-gagement model argues that this is not the case and thatsuch material rewards primarily influence engagementindirectly, by influencing identity status. The key argu-ment of the group engagement model is that people’slevel of cooperation with groups is primarily shaped bythe extent to which they identify with those groups.Cooperation is driven, in other words, by the motiva-tion to create and maintain a favorable identity.

It seemscounterintuitive tomanypeople toargue thatresources are not the primary factor that directly shapesengagement. Certainly, people can think of many exam-ples from their everyday lives that seem to suggest a re-source-based linkagewithengagement.Theseeming im-portance of resource concerns is also supported by someresearch findings. This may reflect evidence of the indi-rectconnectionbetweenresourcejudgmentsandengage-ment of the type we have already outlined. If, as the groupengagement model argues, resource judgments indi-rectly influence engagement, then studies that do notmeasure identity judgments will find a connection be-tween resource judgments and engagement. However,the group engagement model suggests that, in a fullyspecified model, which includes both resource and iden-tity judgments, the spurious connection between re-source judgmentsandengagementwilldisappear(exceptfor that between resources and mandatory cooperation),whereas a mediated connection will remain.

What Organizational ConditionsMatter?

Thegroupengagementmodel alsoconsidershowthepolicies and practices of the group shape identity-based

and resource-based judgments. It is this aspect of thegroup engagement model that directly addresses issuesof justice. The group engagement model argues thatpeople are most strongly influenced by one aspect of thepolicies and practices of their group—the fairness of thegroup’s procedures. This argument builds on the perva-sive finding that procedural justice judgments have astrong and widespread influence on people’s thoughts,feelings, and behaviors in group contexts (Lind & Tyler,1988; Tyler, Boeckmann, Smith, & Huo, 1997; Tyler &Smith, 1997).

In other research, we have argued that the proceduralfairness that influences people in groups emanates fromtwo different sources. First, people are influenced bytheir judgments about the fairness that is linked to theformal rules of the group. These rules can be enshrinedin a constitution or mission statement or articulated bygroup leaders in speeches and written documents. Sec-ond,peoplearealso influencedbytheir judgmentsaboutthe fairness of the implementation of these rules andprocedures by particular authorities (teachers, supervi-sors, parents), by group members (classmates, cowork-ers, siblings) with whom the individual has one-time orongoing personal experiences, or both. Studies suggestthat both formal rules and their implementation by par-ticular authorities have an influence on people’s reac-tions to groups (Rupp & Cropanzano, 2002; Tyler &Blader, 2000).

Further, this research hypothesizes that, for both ofthese two sources of justice—what we term formal andinformal sources—people react to two classes of pro-cedural information. Those classes or categories ofprocedural elements are the quality of the decisionmaking that occurs and the quality of treatment thatpeople experience. Studies suggest that both issues ofdecision making and issues of quality of treatment in-fluence people’s procedural justice-related reactions ingroups (Tyler & Blader, 2000). Note how the inclusionof quality of treatment issues is consistent with thetrend in justice research that we discussed earlier.

By combining the distinction between formal and in-formal sources of justice with the distinction betweendecision making and treatment concerns, we arrive at afour-component model of procedural justice (Blader &Tyler, 2003a, 2003b; Tyler & Blader, 2000). Thefour-componentmodelargues that eachof the fourcom-ponents represents a distinct justice evaluation made bygroup members and, furthermore, that each has a uniqueinfluence in determining overall evaluations of proce-dural justice. Empirical research confirms these asser-tions (Blader & Tyler, 2003a; Tyler & Blader, 2000).

This work on the four-component model links to ourwork on the group engagement model through the hy-pothesis that these procedural elements are importantbecause they shape people’s identities. The group en-gagement model argues that, when people want to makeidentity assessments, one aspect of their lives that they

355

JUSTICE, IDENTITY, AND BEHAVIOR

look to is the procedures of the groups to which they be-long. The group engagement model argues that peoplefocus heavily on issues of the procedural justice of theirgroups because they find procedural justice informationto be the most useful identity-related information theycanhaveabout theirgroups.The four-componentmodelmore specifically stipulates what aspects of procedurespeople use to make judgments about their group-basedidentity.

As we have already noted, it is also possible to con-ceptualize a relationship between the person and thegroup that is centered around the exchange of resources.If this were the key motivation that shapes people’s en-gagement in groups, we would expect that the elementof group policies and practices that would most shapetheir engagement is their estimate of the degree to whichthe rules and policies of the group provide them with de-sirable resources. Such desirable resources may be con-ceptualized as either outcomes that are fair or that are fa-vorable, and thuseitherof these twooutcome judgmentscould affect group members’ resource judgments.These resource judgments could, in turn, influence theirengagement in the group. In the case of either outcomefairness or outcome favorability, it is the concern overthe outcomes that are being received from the group thatwould be driving engagement in groups.

We therefore have stipulated separate antecedentsof the two types of judgments that we recognized ear-lier as being potentially important in determining en-gagement in groups. That is, earlier we described bothresource and identity-based antecedents of engage-ment in groups. Now we have identified the antecedentorganizational conditions that can determine these twoevaluations.

What Is Identity?

As noted earlier, the group engagement model dis-tinguishes among three aspects of identity: identifica-tion, pride, and respect (Tyler & Blader, 2000, 2001).One approach to identity is to define it as identifica-tion, or as the degree to which people merge theirsense of self with the group—thinking of themselvesand the group in similar terms and defining them-selves in terms of their group membership. In Tylerand Blader (2000) we referred to this process ofmerger of self and group as psychological engage-ment in the group. It has also been referred to asidentification with the group. The group engagementmodel hypothesizes that when people identify morestrongly with a group, they will be more willing toact cooperatively in that group—investing their timeand energy in working to see the group succeed.

A second way to conceptualize the role of identityin shaping engagement in groups is to consider the in-fluence of status judgments. If groups serve an impor-tant function in the creation and maintenance of a posi-

tive identity, then group-related status judgmentsshould shape engagement in the group. People shouldbe more engaged in groups that have positive identityimplications for the self, both because association withthe group builds positive identity and because associa-tion is needed to maintain the viability of the group thatsustains that identity.

We conceptualize status assessments as being re-flected by two different concepts—pride and respect.Pride reflects judgments about the status of the group,also indexed by measures of group prestige (Mael &Asforth, 1992; Smidts, Pruyn, & val Riel, 2000). It ex-presses a person’s view about the status of their group.People who belong to groups that they feel have highstatus feel good about themselves by virtue of their as-sociation with the group. These feelings stem primarilyfrom noncomparative feelings of inclusion in a highstatus group, rather than from comparisons of one’sgroup to other groups (Tyler & Blader, 2002).

Respect reflects judgments about one’s status withinthe group. It expresses a person’s view about their statusin the eyes of other group members. Respect is also re-ferred to as social reputation (Emler & Hopkins, 1990).Whereas social identity theory was originally focusedon the status of groups (i.e., on intergroup phenomena),it is also recognized that people are influenced by theirjudgments of their status within groups (Doosje et al.,1999; Noel, Wann, & Branscombe, 1995).

Research supports the premise of the group engage-ment model by showing that identification, pride, andrespect are connected to feelings of self-esteem andself-worth (Tyler & Blader, 2000; Tyler, Degoey, &Smith, 1996). This is consistent with the argument ofthe group engagement model that people use groupidentity-based judgments to evaluate themselves.

These three identity elements of the group engage-ment model—identification, pride, and respect—areeach predicted to be related to engagement in groups.However, they are not equivalent constructs. In partic-ular, status indicators (pride, respect) are hypothesizedto influence identification, as shown in Figure 1. Thisrelationship develops because people should be morehighly motivated to merge their identity with a groupwhen the group has high status (pride), when they feelthat they have status in the group (respect), or both.Doing so ensures that their identification is with agroup that makes them feel good about their socialselves, with the obvious benefits that that entails fortheir overall sense of self. Pride and respect, in otherwords, engender identification with the group in peo-ple’s motivated attempts to develop and maintain apositive social identity.

The Group Engagement Model

To summarize, we have developed several argu-ments regarding the antecedents of group engagement.

356

TYLER AND BLADER

First, we have outlined two potential proximal determi-nants of group members’engagement: (a) their identityjudgments and (b) their resource judgments. Regard-ing these proximal determinants, we have argued thatidentity judgments are the central issue for predictingengagement and cooperation, and we have further ar-gued that the influence of resource judgments is medi-ated by these identity concerns. Further, we describedthese identity judgments as composed of both statusand identification concerns. Second, we have identi-fied the organizational conditions that are linked tothese two potential determinants of engagement. In thecase of identity evaluations, we have argued that proce-dural justice is a key antecedent; group members’ iden-tities vis-à-vis the group are determined in large part bytheir evaluations of the fairness of the group’s proce-dures. These process fairness evaluations, in turn, arecomposed of four distinct and important judgments. Inthe case of resource evaluations, we have argued thatissues of both outcome fairness (i.e., distributivejustice) and outcome favorability are key antecedents.

This set of hypotheses leads to an integrated model ofwhat leads group members to engage in theirgroups—the group engagement model, presented inFigure 1. The overall group engagement model makesclearwhypeople focussoheavilyonwhetherornot theirgroups’ procedures are fair. The procedural fairnessjudgment provides key information that shapes the de-gree to which people regard their group as having highstatus, regard themselves as having high status in theirgroup, and identify with the group by merging theirsense of self with the group. Procedural justice judg-mentsare thusakeyantecedentof identityassessments.

Identity assessments, in turn, are the key determinantof important psychological and behavioral connectionsto the group. The degree to which people identify withtheir group shapes the degree to which they develop sup-portive attitudes and values and the degree to which theyengage themselves behaviorally in the group.

Outcome judgments, such as outcome fairness andoutcome favorability, do influence evaluations of theresources received from the group. However, our groupengagement model argues that these resource judg-ments do not directly influence attitudes, values, or be-havioral engagement. Instead, it is argued that their in-fluence on these important group outcomes ismediated by group members’ identity judgments.

Are the proposals of the group engagement modelvalid? Tyler and Blader (2000) provided a preliminarytest of the model using survey data from 404 employeesdrawn from a variety of work organizations. Usingcausal modeling, they tested several of the key hypothe-ses of the group engagement model and found supportfor all of them (see Tyler & Blader, 2000, p. 196). First,they found that identity judgments shaped attitudes, val-ues, and cooperative behaviors. Consistent with the pre-dictions of the model, they found a greater influence of

identity judgments on discretionary, as compared tomandatory, behavior. Second, resource judgments werefound to influence attitudes, values, and discretionarycooperative behaviors indirectly, through identity judg-ments, but not directly. Third, procedural justice judg-ments were found to be the primary antecedent of iden-tity judgments (Tyler & Blader, 2000, p. 136). Therelationship between pride, respect, and identificationwas not examined in this initial test of the model, andthus awaits empirical confirmation.

Trends in Justice Research and theGroup Engagement Model

When we reconsider the trends in the justice litera-ture that were discussed earlier, we find that they areconsistent with the arguments of the group engagementmodel. First, research attention has shifted from anearly exclusive focus on the influence of people’s judg-ments about distributive justice to a more recent focuson the influence of both distributive and proceduraljustice judgments. This is consistent with the group en-gagement model’s assertion that procedural justicejudgments are central antecedents of how people de-velop their identities in relation to their groups. Sec-ond, within the study of procedural justice, researchhas shifted from exclusively defining procedural fair-ness by the quality of decision-making procedures tobroader definitions of procedural fairness that alsoconsider the quality of people’s interpersonal treat-ment when they are interacting with others. This isconsistent with the prominent role that treatment crite-ria of procedural justice play in the group engagementmodel, and in the clear linkage of this class of criteriawith concerns about group-related identity.

Third, there has been a shift away from focusing pri-marily on negative reactions to experiences—anger, dis-satisfaction, and negative behaviors ranging from riotingto sabotage (Crosby, 1976; Greenberg, 1990)—towardgreater attention to positive attitudes and values and co-operative behaviors (Tyler & Blader, 2000). This is re-flected in the breadth of attitudes and behaviors addressedby the group engagement model. Such a broader focus isconsistent with the idea that justice models should and doprovide a general framework within which we can under-stand people’s connections to groups.

Thus, we argue that the underlying psychologicaldynamics suggested by the group engagement modelcan shed light on how and why justice research hasevolved in many of the respects that it has. Whereasjustice researchers’ initial attention was focused on is-sues that reflected the context in which they were oper-ating, empirical results suggested alternative ap-proaches that provided relatively better insight into theissue of group engagement. Those empirical findingscan be understood by considering the essential argu-ments put forth by the group engagement model.

357

JUSTICE, IDENTITY, AND BEHAVIOR

Implications of the Group EngagementModel

In the remainder of this article, we further exploreseveral issues that flow from the theoretical frameworkput forth by the group engagement model. Specifically,we consider the fundamental issue of why proceduraljustice demonstrates the important links to identity thatwe predict and observe. Further, we consider some im-portant distinctions between the two status judgmentswe have outlined (pride and respect). These issues pro-vide further insight into the psychological underpin-nings of the model, and specify important new direc-tions for empirical research on the model.

Procedural Justice as Identity Security

Why does procedural justice play such an importantrole in shaping identification with the group, which inturn links procedural justice with cooperation in thegroup? We propose that procedural justice providesidentity security. A merger of the self with the groupmay provide people with support for positive feelingsof self-worth and high self-esteem, through their con-nection to the group. By being members of a group,people can first use the group as a source of iden-tity-relevant categories through which they definethemselves. In addition, they can use the status of thegroup as a source of self-affirmation—gaining confi-dence in their own identity through their associationwith the group. Thus, people have a great deal to gainby their association with groups, at least if that associa-tion has favorable identity implications.

Although using a group to determine one’s identitycan facilitate positive feelings of self-worth and self-es-teem, it also contains risks. People can receive favorableidentity-relevant information, but they can also havetheir identities damaged when they receive negativefeedback from the group. So, for example, a person whoidentifies more strongly with a group is more psycho-logicallydamagedwhen theysee that thegroupoperatesin negative ways (Brockner, Tyler, & Cooper-Schnei-der, 1992). Opening one’s self up to the group createsvulnerabilities and opens up the possibility of receivingnegative feedback that damages one’s identity. Peopleare sensitive to the potential pitfalls of identificationwithgroups thatprovide themwithnegativestatus infor-mation. Status evaluations (pride, respect) positivelypredict identification with the group, and thus the morefavorable these status evaluations, the stronger people’slevel of identification with the group.

Whereas the risk of receiving damaginginformation exists for everyone, it is especially strongfor those members of groups who are vulnerable. Peo-ple are more vulnerable when their status in the groupis unclear, when it is unclear whether they are includedin the group (Tyler & Lind, 1990), or both. When asso-

ciation with a group is less unclear but instead clearlynegative, deleterious consequences are even morelikely. For instance, people who are members of certainstigmatized demographic groups often have negativestereotypes applied to them, making them feel low sta-tus within the broader superordinate groups of whichthey are also members. These stereotypes can have aserious impact on these individuals. Research inachievement settings, for example, demonstrates thatthe behavior of people who are potentially vulnerableto stereotype application is changed, with people reluc-tant to engage themselves psychologically andbehaviorally in tasks that might result in identity dam-aging feedback linked to confirming negative sub-group stereotypes (Steele, Spencer, & Aronson, 2002).

The degree to which people identify reflects their ef-fort to balance the potential identity gains associatedwithmerging their identitieswithagroupagainst thepo-tential risksof that samemergerof theselfand thegroup.The group engagement model argues that, to the degreethat people feel that the group makes decisions via fairprocedures, theyaremore likely to feel that their identitycanbesafelyandsecurelymergedwith thatof thegroup.Procedural justice, in other words, appears to allay peo-ple’s concerns that group membership will result in neg-ative consequences for the self; it provides them with asense of identity security. The existence of such securityleads people to feel comfortable engaging psychologi-cally and behaviorally in groups.

This argument only raises a more fundamentalquestion, however. Why does procedural justice pro-vide identity security? Because each of the two func-tions of procedures—quality of decision making andquality of interpersonal treatment—contributes to peo-ple’s assessment that it is safe for them to merge theiridentity with their group. Consider the quality of thedecision making in the group. If a group makes deci-sions unfairly (i.e., inconsistently, based on biases orpersonal opinions instead of facts), then there is a riskthat stereotypes or personal prejudices might poten-tially be applied to group members that belong to par-ticular subgroups. Experiencing stereotyping and prej-udice within the groups that people belong to isdamaging to their sense of self, which may in turn leadthem to maintain a psychological distance betweentheir identity and group membership. Fair proceduresreassure people that stereotypes are not and will not beapplied. Neutrality, in other words, encourages confi-dence in the security of including the group in one’ssense of self, leading people to be more willing to en-gage in a group.

To understand how this argument differs from a so-cial exchange perspective, consider the ex-change-based procedural justice model proposed byThibaut and Walker (1975). This model argues thatpeople’s motivation in disputes with others is to gainfair outcomes—that is, the resources they deserve.

358

TYLER AND BLADER

They value the opportunity to present evidence (pro-cess control) because it allows them to shape the out-come of a third parties’ decisions (decision control),leading that third party to make a decision consistentwith principles of distributive justice. Hence, the goalof a procedure is to provide outcome security by pro-ducing a fair distribution of resources. The group en-gagement model is fundamentally different; it pro-poses that it is identity security that is key.

An example of the importance of fair decision-mak-ing procedures is provided by judgments about racialprofiling (Tyler & Wakslak, 2002). When people arestopped by the police, they must infer whether they arebeing stigmatized by legal authorities via the applica-tion of a negative group stereotype to them based not onwhat they are doing, but on their race, gender, or age. Ifthey are, this behavior by group authorities carries nega-tive identity implications, raising questions aboutwhether they are included in the rights accorded tomembers of the superordinate group (i.e., the rights ac-corded to group members in good standing). What reas-sures people that profiling is not occurring, so that theticket they have received does not reflect negatively ontheir social status? If people infer that the authorities aremaking their decisions fairly, they are less likely to saythat they are being profiled. These inferences and theidentity implications that flow from them may havelarge consequences on these group members’ attitudes,values, and cooperation with the superordinate group.

The second aspect of procedural justice is the qual-ity of the treatment that a person experiences whendealing with others. This aspect of procedural justicefeeds directly into people’s identity judgments, be-cause treatment with dignity and politeness, as well asthe consideration of one’s needs and concerns, are alsoaspects of interpersonal experience that communicatethat one is valued by others. Again using the exampleof racial profiling, studies suggest that people are lesslikely to think that they have been profiled when theyare treated politely and with dignity, when they feelthat their needs and concerns are recognized, and whenthey feel that their rights are acknowledged. People in-fer that police officers who treat them politely and withrespect are affirming their status rather than undermin-ing or raising questions about it. Hence, quality oftreatment is also associated with whether people feelthat stereotypes are being applied. In addition, evalua-tions that the treatment experienced in the group con-text is fair reassures people that they will receive treat-ment that affirms their status well into the future oftheir group membership. This also reassures people’ssense of identity security.

The example of racial profiling illustrates the risks aperson undertakes when merging one’s sense of selfinto a group. If people are drawing their sense of selffrom a superordinate group membership, then demean-ing and disrespectful treatment from that superordinate

group will undermine their feelings of favorableself-esteem and self-worth. It will communicate mar-ginality and exclusion from important protections thatare extended to most other group members—for exam-ple, “freedom from arbitrary arrest and seizure.” Racialprofiling illustrates the reality that group members arenot able to prevent stereotype application in all set-tings. People are not able to avoid being in situations inwhich they must contend with the police, the courts,and other group authorities (Lipsky, 1980).

The group engagement model argues that each ofthese two aspects of the procedural justice experiencedin a group carries important identity information to thegroup member, with the result that procedural justicejudgments are especially important to people whenthey are evaluating groups. Procedural justice is a par-ticularly important source of this information, we ar-gue, because the phenomena of process fairness pro-vides people with reassurance that they can safely drawa significant portion of their sense of self from thegroup. Procedures communicate information that peo-ple care about in evaluating their group memberships,and furthermore allow them to make inferences aboutthe nature of their future connection to the group. Suchinferences are important for determining their engage-ment in the group.

Pride Versus Respect

This discussion has treated pride and respect as dis-tinct but related aspects of people’s status judgments.However, it is possible to distinguish between thesetwo aspects of status and make distinct predictionsabout them. In particular, pride reflects the categoricalself, whereas respect reflects the reputational self (Ty-ler & Smith, 1999). Thus, these two aspects of statusspeak to different psychological drives: the drive tohave a positive social identity and the drive to have apositive personal identity.

The categorical self reflects a focus on category at-tributes and in particular on the status of thegroup—what is good or bad about the group, its values,its identity, and so on. It diminishes distinctions amonggroup members and focuses on common attributes.Hence, the categorical self is created from prototypicalelements of the group. This focus leads to a motivationto be loyal to the group, its values, its rules, and its au-thorities. Hence, we predict it will be linked to publicbehaviors that display loyalty. We hypothesize that itwill lead to increased attention to the group and groupvalues, leading to conformity and uniformity of behav-ior among group members.

The reputational self is based on a focus on the per-son within the group. People’s interest in how others inthe group view them leads to attention to their uniqueand valuable attributes, as those attributes are identi-fied by them and by others in the group. The

359

JUSTICE, IDENTITY, AND BEHAVIOR

reputational self reflects our view of ourselves in theeyes of others. This focus leads to unique and creativeactions on the part of group members, actions designedto create reputational capital by enhancing their favor-able image among members of the group. People whofeel respected by others in their groups are predicted tobecome highly committed to the group and voluntarilymotivated to act in ways that make use of distinctivequalities and abilities.

The group engagement model thus suggests thatpride will be particularly linked to mandatory–requiredbehavior, whereas respect will be linked especiallystrongly to discretionary–voluntary behavior. Theseinfluences are expected in addition to the indirect in-fluence that pride and respect have via identification.This prediction grows out of the recognition that man-datory behaviors will likely be highly prototypicalgroup behaviors, since they are identified and stipu-lated by the group as desirable. Discretionary behav-iors, on the other hand, originate with the individualand are thus more idiosyncratic in nature, and shouldthus be related to that aspect of the self that strives toexpress and protect one’s individuality. The modelmakes the additional predictions that decision makingwill be a particularly important antecedent of pride andquality of treatment of respect. These predictions, ofcourse, are in need of empirical testing.

Summary

Either an identity or a resource model could poten-tially explain people’s level of engagement in groups.However, the group engagement model suggests that itis identity motivations that are key to engagement.Hence, this highlights the need to understand whatshapes people’s identities. The group engagementmodel argues that it is procedural justice that is centralto how and whether people construct their group-re-lated identities.

These arguments of the group engagement modelmake clear why justice research has shifted from a fo-cus on distributive justice to a focus on procedural jus-tice and, within procedural justice, from a focus on de-cision making to a focus on both decision making andquality of treatment. These shifts have occurred be-cause both assessments of the quality of decision mak-ing and assessments of the quality of treatment providepeople with the most useful information in construct-ing and maintaining their identities.

Compelling evidence of this centrality of proceduralissues to identity is provided by the importance ac-corded to the quality of treatment people experiencewhen dealing with others. Quality of treatment provideslittle informationabout the favorabilityor fairnessof theoutcomes being received, yet it is consistently found tobe a key input into procedural fairness judgments and an

important factor shaping engagement in the group. It isdifficult to understand this centrality without accordingidentity an important role in mediating the impact ofprocedural justice on engagement. Although we arguethat decision-making processes are also evaluated froman identity perspective, their linkages to allocationsthemselves makes them less conclusive indicators of theimportance of identity issues.

By reviewing research on social justice from theperspective of the theoretical framework representedby the group engagement model, we have attempted toprovide a coherent way of understanding the changesthat have occurred within the field over the last severaldecades. We believe that this framework suggests thatthe findings of social justice research speak to socialpsychology more broadly. That is, the core implicationof the group engagement model about what matters topeople in social contexts has relevance not only to so-cial justice research, but to work on group dynamicsand social psychology as a field.

References

Alexander, S., & Ruderman, A. (1986). The role of procedural anddistributive justice in organizational behavior. Social JusticeResearch, 1, 177–198.

Bies, R. J., & Moag, J. S. (1986). Interactional justice. In R. J.Lewicki, B. M. Sheppard, & M. H. Bazerman (Eds.), Researchon negotiations in organizations (Vol. 1, pp. 43–55). Green-wich, CT: JAI.

Blader, S. L., & Tyler, T. R. (2003a). A four component model ofprocedural justice: Defining the meaning of a “fair” process.Human Resource Management Review, 13, 107–126.

Blader, S. L., & Tyler, T. R. (2003b). What constitutes fairness inwork settings? A four component model of procedural justice.Personality & Social Psychology Bullentin, 29, 747–758.

Brockner, J., Tyler, T.R., & Cooper-Schneider, R. (1992). The influ-ence of prior commitment to an institution on reactions to per-ceived unfairness: The higher they are, the harder they fall. Ad-ministrative Science Quarterly, 37, 241–261.

Crosby, F. (1976). A model of egoistical relative deprivation. Psy-chological Review, 83, 85–113.

Dansereau, F., Graen, G., & Haga, W. J. (1975). A vertical dyad link-age approach to leadership within formal organizations. Orga-nizational Behavior and Human Performance, 13, 46–78.

Doosje, B., Ellemers, N., & Spears, R. (1999). Commitment and in-tergroup behavior. In N. Ellemers, R. Spears, & B. Doosje(Eds.), Social Identity (pp. 84–106). London: Blackwell.

Emler, N., & Hopkins, N. (1990). Reputation, social identity, and theself. In D. Abrams & M. A. Hogg (Eds.), Social Identity Theory(pp. 113–130). New York: Springer-Verlag.

Eisenberger, R., Huntington, R., Hutchinson, S., & Sowa, D. (1986).Perceived organizational support. Journal of Applied Psychol-ogy, 71, 500–507.

Foa, U. G., & Foa, E. B. (1974). Societal structures of the mind.Springfield, IL: Charles C. Thomas.

Greenberg, J. (1990). Employee theft as a reaction to underpaymentinequity. Journal of Applied Psychology, 75, 561–568.

Greenberg, J., & Folger, R. (1983). Procedural justice, participation,and the fair process effect in groups and organizations. In P. B.Paulus (Ed.), Basic group processes (pp. 235–258). New York:Springer-Verlag.

360

TYLER AND BLADER

Gurr, T. (1970). Why men rebel. Princeton, NJ: Princeton UniversityPress.

Hogg, M. A., & Abrams, D. (1988). Social identifications. London:Routledge.

Kelley, H. H., & Thibaut, J. W. (1978). Interpersonal relations: Atheory of interdependence. New York: Wiley.

Levine, R. A., & Campbell, D. T. (1972). Ethnocentrism: Theories ofconflict, ethnic attitudes, and group behavior. New York: Wiley.

Lind, E. A., Kanfer, R., & Earley, C. (1990). Voice, control, and pro-cedural justice. Journal of Personality and Social Psychology,59, 952–959.

Lind, E. A., & Tyler, T. R. (1988). The social psychology of proce-dural justice. New York: Plenum.

Lipsky, M. (1980). Street-level bureaucracy. New York: Rus-sell-Sage.

Locke, E. A., & Latham, G. P. (1990). A theory of goal setting andtask performance. Englewood Cliffs, NJ: Prentice-Hall.

Mael, F. A., & Ashforth, B .E. (1992). Alumni and their alma mater.Journal of Organizational Behavior, 13, 103–123.

Messick, D. M., Bloom, S., Boldizar, J. P., & Samuelson, C. D.(1985). Why we are fairer than others. Journal of ExperimentalSocial Psychology, 21, 389–399.

Messick, D., & Sentis, K. (1985). Estimating social and nonsocialutility functions from ordinal data. European Journal of SocialPsychology, 15, 389–399.

Mikula, G., Petri, B., & Tanzer, N. (1990). What people regard as un-just. European Journal of Social Psychology, 22, 133–149.

Montada, L. (1995). Applying social psychology: The case of redistri-butions in united Germany. Social Justice Research, 8, 73–90.

Nagin, D. (1998). Criminal deterrence research at the outset of thetwenty-first century. In M. Tonry (Ed.), Crime and justice: A re-view of research (Vol. 23, pp. 1–42). Chicago: University ofChicago Press.

Noel, J. G., Wann, D. L., & Branscombe, N. R. (1995). Peripheralingroup membership, status and public negativity towardsoutgroups. Journal of Personality and Social Psychology, 68,127–137.

Ross, M., & Sicoly, F. (1979). Egocentric biases in availability andattribution. Journal of Personality and Social Psychology, 37,322–336.

Rupp, D. E., & Cropanzano, R. (2002). The mediating effects of so-cial exchange relationships in predicting workplace outcomesfrom multifoci organizational justice. Organizational Behaviorand Human Decision Processes, 89, 925–946

Rusbult, C., & Van Lange, P. (1996). Interdependence processes. InE. T. Higgins & A. W. Kruglanski (Eds.), Social psychology(pp. 564–596). New York: Guilford.

Sedikides, C., & Brewer, M. B. (2001). Individual self, relationalself, and collective self. Philadelphia: Psychology Press.

Smidts, A., Pruyn, A. T. H., & van Riel, C. B. M. (2000, July). Man-aging organizational identification through employee commu-nication. Paper presented at the EAESP Small Group meetingon social identity processes in organizations, Amsterdam, TheNetherlands.

Snyder, C. R., & Lopez, S. J. (2002). Handbook of positive psychol-ogy. New York: Oxford.

Steele, C., Spencer, S. J., & Aronson, J. (2002). Stereotype threat, in-tellectual identification, and academic performance of womenand minorities. In M. Zanna (Ed.), Advances in experimentalsocial psychology (Vol. 34, pp. 379–440). New York: Aca-demic.

Thibaut, J., & Kelley, H. H. (1959). The social psychology of groups.New York: Wiley.

Thibaut, J., & Walker, L. (1975). Procedural justice. Hillsdale, NJ:Lawrence Erlbaum Associates, Inc.

Thompson, L., & Loewenstein, G. (1992). Egocentric interpretationsof fairness and interpersonal conflict. Organizational Behavior& Human Decision Processes, 51, 176–197.

Tyler, T. R. (1987). Conditions leading to value expressive effects injudgments of procedural justice: A test of four models. Journalof Personality and Social Psychology, 52, 333–344.

Tyler, T. R. (1988). What is procedural justice? Criteria used by citi-zens to assess the fairness of legal procedures. Law and SocietyReview, 22, 103–135.

Tyler, T. R. (1989). The psychology of procedural justice: A test ofthe group value model. Journal of Personality and Social Psy-chology, 57, 830–838.

Tyler, T. R. (1994). Psychological models of the justice motive. Jour-nal of Personality and Social Psychology, 67, 850–863.

Tyler, T. R. (1999). Why people cooperate with organizations: Anidentity-based perspective. Research in Organizational Behav-ior, 21, 201–246.

Tyler, T. R. (2000). Social justice: Outcome and procedure. Interna-tional Journal of Psychology, 35, 117–125.

Tyler, T. R., & Bies, R. (1990). Interpersonal aspects of proceduraljustice. In J. S. Carroll (Ed.), Applied social psychology in busi-ness settings (pp. 77–98). Hillsdale, NJ: Lawrence Erlbaum As-sociates, Inc.

Tyler, T. R., & Blader, S. L. (2000). Cooperation in groups: Proce-dural justice, social identity, and behavioral engagement. Phil-adelphia: Psychology Press.

Tyler, T. R., & Blader, S. L. (2001). Identity and cooperative behav-ior in groups. Group Processes and Intergroup Relations, 4,207–226.

Tyler, T. R., & Blader, S. L. (2002). Autonomous vs. comparativestatus: Must we be better than others to feel good about our-selves? Organizational Behavior and Human Decision Pro-cesses, 89, 813–838.

Tyler, T. R., Boeckmann, R., Smith, H. J., & Huo, Y. J. (1997). Socialjustice in a diverse society. Denver, CO: Westview.

Tyler, T. R., & Caine, A. (1981). The role of distributional and proce-dural fairness in the endorsement of formal leaders. Journal ofPersonality and Social Psychology, 41, 642–655.

Tyler, T. R., Degoey, P., & Smith, H. (1996). Understanding why thejustice of group procedures matters: A test of the psychologicaldynamics of the group-value model. Journal of Personality andSocial Psychology, 70, 913–930.

Tyler, T. R., & Huo, Y. J. (2002). Trust in the law. New York: Rus-sell-Sage.

Tyler, T. R., & Lind, E. A. (1990). Intrinsic versus community-basedjustice models. Journal of Social Issues, 46, 83–94.

Tyler, T. R., & Lind, E. A. (1992). A relational model of authority ingroups. Advances in Experimental Social Psychology, 25,115–191.

Tyler, T. R., Rasinski, K., & Spodick, N. (1985). The influence ofvoice on satisfaction with leaders: Exploring the meaning ofprocess control. Journal of Personality and Social Psychology,48, 72–81.

Tyler, T. R., & Smith, H. J. (1997). Social justice and social move-ments. In D. Gilbert, S. Fiske, & G. Lindzey (Eds.), Handbookof social psychology (4th ed., Vol. 2, pp. 595–629). New York:McGraw-Hill.

Tyler, T. R., & Smith, H. J. (1999). Justice, social identity, and groupprocesses. In T. R. Tyler, O. John, & R. Kramer (Eds.), The psy-chology of the social self. Hillsdale, NJ: Lawrence Erlbaum As-sociates, Inc.

Tyler, T. R., & Wakslak, C. (2002). Profiling and the legitimacy ofthe police: Procedural justice, attributions of motive, and theacceptability of social authority. Unpublished manuscript, NewYork University.

van den Bos, K., & Lind, E. A. (2002). Uncertainty management bymeans of fairness judgments. In M. P. Zanna (Ed.), Advances inexperimental social psychology (Vol. 34, pp. 1–60). New York:Academic.

Walster, E., Walster, G. W., & Berscheid, E. (1978). Equity: Theoryand research. Boston: Allyn and Bacon.

361

JUSTICE, IDENTITY, AND BEHAVIOR